ASCENT: Online Test-Time Training of Long-Horizon Agents via Self-Distillation of Verified Experience

arXiv:2610.05303 · cs.LG · Submitted 2026-10-04 · Read on arXiv

cs.LG

Submitted: 2026-10-04

Updated: 2026-10-04

Terminology

• Distillation — 1× in this paper · explained in 2 episodes
• Self-Distillation — 1× in this paper · explained in 7 episodes

Sources

• EvoSkill: Automated Skill Discovery for Multi-Agent Systems
• CLaaS: Continual learning as a service for sample efficient online learning
• Reinforced Self-Training (ReST) for Language Modeling
• Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning
• REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization
• WHALE: A Simple Recipe for Joint Harness-Weight Optimization
• When Privileged Guidance Misaligns: State-Matched Routing and Contextualized Self-Distillation for Multi-Turn Agents
• Self-Distilled Agentic Reinforcement Learning
• Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills
• Scaling Self-Evolving Agents via Parametric Memory
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Hindsight Credit Assignment for Long-Horizon LLM Agents
• TTPO: Test-Time Policy Optimization
• Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents
• No Time Like the Present: Agentic Test-Time Training for LLM Agents
• AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning
• SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning
• HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents
• Scaling Relationship on Learning Mathematical Reasoning with Large Language Models
• Reinforcing Multi-Turn Reasoning in LLM Agents via Turn-Level Reward Design

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks